PURPOSE: In clinical research, understanding of the risks and benefits of participation is a prerequisite of informed consent. We previously found that many patients enrolled in early-phase oncology trials report high expectations of benefit from participation. Also, some patients are unable or unwilling to predict their outcomes. We sought to evaluate whether such responses reflect poor understanding or other factors.

METHODS: Forty-five patients who had enrolled in phase 1 or 2 oncology clinical trials participated in this study. Participants were randomized to 1 of 3 interview protocols corresponding to 3 questions about likelihood of benefit: frequency-type ("Out of 100 patients who participate in this study, how many do you expect will have their cancer controlled as a result of the experimental therapy?"); belief-type ("How confident are you that the experimental therapy will control your cancer?"); and vague ("What is the chance that the experimental therapy will control cancer?"). In cognitive interviews, we asked participants about how they understood and answered the target question. Each participant then answered and discussed one of the other target questions. Only response justifications from the first target response were included in this analysis.

RESULTS: A total of 19 participants (42%) either could not answer the first target question or justified their estimate with a version of "don't know." This result did not differ by question type (P=.46). Versions of "don't know" justifications included: question is unanswerable, because no one has studied the therapy before (n=6); uncertainty about the precision of the estimated likelihood of benefit (n=2); patient feels unqualified to answer (n=2); patient reports having insufficient information about the therapy or other participants to answer (n=1); and patient has not considered the likelihood of benefit, so has no expectation (n=1).

CONCLUSIONS: Careful probing of patients' responses revealed different reasons for being unable to provide an expectation of benefit in early-phase trials. However, not all types of "don't know" responses reflected a lack of understanding. For example, many patients could not provide an answer because they noted correctly that no one knows the chance of benefit. In the context of uncertain outcomes, patients who cannot or refuse to provide an answer must be queried further to determine whether their responses should raise concern about the quality of informed consent.